Skip to Content
MIT Technology Review

How to make sense of all the information about coronavirus

We talked to the author of a viral explainer on coronavirus about how to find and communicate trustworthy information in the midst of a pandemic.

April 3, 2020

Note: This episode has ended.

In this episode of Radio Corona, Gideon Lichfield, editor in chief of MIT Technology Review, spoke with Tomas Pueyo, whose Medium post “Coronavirus: Why You Should Act Now” has become one of the defining explainers on the internet about the coronavirus outbreak (it has been viewed more than 40 million times, and translated into at least 30 languages).

In interviews, Pueyo is quick to point out he is not an epidemiologist. He is the vice president of growth at Course Hero, an online learning platform. Even so, his post synthesized data available about the outbreak into a compelling and clear argument that influenced many people's thinking.

This episode was recorded on April 3, 2020. You can watch it below.

Gideon Lichfield: I'm Gideon Litchfield. I'm the editor in chief of MIT Technology Review. Thank you for joining us on this latest edition of Radio Corona, our show about the coronavirus and how it's impacting the world. And I'm joined by Tomas Pueyo who is a VP of growth at Course Hero, which is online learning platform, but has somehow turned into the hero of explaining the coronavirus pandemic of the last few weeks. He has written a series of posts on medium, three of them, the latest, I think published today, and they explain the coronavirus pandemic in only one of the clearest ways that I've seen. And I'm reading a lot, a lot of what is on the internet about coronavirus. They are long, they're full of charts, stories that are full of charts don't always become things that are viral on the internet.

Tomas Pueyo: Don't usually, yeah.

Gideon Lichfield: These things that become very, very viral. Tomas, the first one that you wrote "Coronavirus: why you must act now." How many views is that up to now?

Tomas Pueyo: I haven't checked in a couple of weeks, but Oh, I mean a week, but, it passed the 40 million, four zero last week.

Gideon Lichfield: So for context that is on the same order of magnitude as the traffic that MIT Technology Review's website gets in an entire year or a smaller website, but still it's pretty impressive. So it means a lot of people have been reading these posts, and I assume that they've been influencing people's understanding of the pandemic, uh, and hopefully making them see the gravity of it in ways that they might not have seen before. So what I want to do is start by talking about how you came to be doing this, you know, tell us a little bit about yourself and then how you ended up becoming a pandemic writer.

Tomas Pueyo: Really by accident. I've been writing for over a year and a half on medium and some of the articles had gone somewhat viral, but nothing like this. Since I write blog posts, I like going deep into different problems, understanding them and explaining them. And, around now a month and a half ago when coronavirus was starting to pop out outside of China, I started looking into the data. And I posted a couple of things on Facebook and surprisingly a lot of people reacted to it like nothing I had seen before. And they were starting to ask me questions. And so I just kept going. I kept looking at the data, kept trying to understand it and shared it with my friends and I started posting every day. I did that for around two weeks. And what I was seeing is a huge need, from very different people across the board, to understand it better.

One of the key needs for example I was seeing is businesses didn't know what to do about it. For sure governments, but businesses should we close? We not? Should we work from home? when? why? And so it was very much meant to address these questions every day until at some point somebody told me, Hey, what you've been writing is very helpful here in Paris. I have a group of friends who would benefit from knowing these things and can you, can you write something for them in posts? I just summarized everything. Can you hear me? Yeah. I just summarized everything, that I had written for a couple of weeks into one post and it just became viral.

Gideon Lichfield: You also cite various people that you relied on for support and help. I mean, how much of this were you putting together by consulting with experts or other source of data?

Tomas Pueyo: The first article I just wrote by myself, but obviously interacting with the community all the time in my Facebook post to understand what they needed and get feedback from them and hear their points and their arguments thanks to that group of volunteers, joined me, especially to do research because there's so much that needs to be analyzed and it's very, very hard to do. So that was a team effort, probably a dozen people, gathering all the info, modeling everything and I've been, I've been working with them ever since.

Gideon Lichfield: And these are all volunteers?

Tomas Pueyo: It's all volunteers and none of them were experts for the second one. And then on the third one, then I started having some experts. I'm working, for example, on a paper with a PhD in epidemiology. I've worked with a Stanford economics professor to get feedback from him. So there's a few experts now that I have access to thankfully, through the exposure that these articles, has gotten.

Gideon Lichfield: Do you have a lot of experts writing into after you published the first two saying, no, you've got this wrong or whatever?

Tomas Pueyo: There's so many interesting things about it. I think the- a lot of people ask me like, who are you and what why should we listen? And the answer to that is, I am nobody and you should not listen to me. And that's why I put all the data there and all the graphs and all the sources so that everybody can not only follow the arguments, but also challenge them. And in some cases, there's been some challenges. And in some cases those challenges have been right. So for example, in the article, I suggest that the speed of virus mutation was very, very, very high. And, I got pushback back from some people say, you're right, it's meeting fast. That speed is not suggesting that there's going to be a complete different strain that requires a different set of different immune response altogether.

And so that's one of the corrections. But I think the, so that's the key. The answer is I'm nobody, don't listen to me. Just listen to the arguments. Use your judgment and so that was interesting. Another thing that I heard a lot was, some, okay, we should be listening to epidemiologists, not to you. And so my answer to that is, which one? Cause you have some that say this, some say this, and it's not very easy. And so thankfully I had a bunch of epidemiologists come to me and say, Hey, actually I learned a lot from you the way you put it. And I created a specific article just with endorsements to, so people might feel more comfortable, listening to my arguments. But to me, I think the best story around this is something that happened on TV in the UK. I was talking with an expert, epidemiologist who was, supporting the government's position. And, at the time, the government's position was we're going to let this run. But they didn't realize that letting this run would kill hundreds of thousands of people.

Gideon Lichfield: Right? This was back when the U.K. said it's better to just let everybody get the virus and get herd immunity and then,

Tomas Pueyo: Yeah, and then, oh, by the way, for the UK, you might kill between 200,000 and 500,000 people. So are you ready to do that? Not only that, but also even if they were going to go for a herd immunity, it meant setting the curve immediately and they were not taking immediate measures. And then 10 days later, the government, pulled back off that strategy and use the more of a suppression strategy of hard measures upfront. And so, that's a very good example. They were making fun of me because I was, I was putting my hands on the, on my face during the, on TV. But, when you're hearing an expert tell everybody, we're doing this and Oh, I didn't realize there's going to be hundreds of people dying, then you need to challenge these people.

Gideon Lichfield: Right. So how do you, by the way, I'll remind anyone who's just joining. We're talking Tomas Pueyo and you can ask questions in the chat. If you're joining us via YouTube, you can throw in questions there as well, and we'll what we'll see them and I will put them to Tomas. The, so one of my colleagues actually asked me a related question, which is lots of people, actually me included, have been on the internet criticizing so-called armchair epidemiologists, who analyze the data and give their take. So you've seemed to manage to get everything right, or at least to have been able to, balance the differing views and present things in a fair way. There was another epidemiologist well, a guy who presented himself as, I don't know if he'd present themselves an epidemiologist, a guy called Aaron Ginn who did a long piece on Medium analyzing why the response to coronavirus was overblown. And Medium ultimately took the post down as misinformation because he got so many things wrong and he was being criticized so hard by so many places. So I guess the question is like, how did you, especially given that there are so many conflicting, models, some projections, and views, how did you balance them out and make sure that you weren't presenting anything that was misinformation or that was just wrong?

Tomas Pueyo: Yeah. So I think, the reason why the article went viral is complete luck. That you can't predict these things. And, there's a lot of requirements to get viral. But then on top of that, we need a lot of luck. So, I think, um, some of the, there are some experience I've had in the past that have helped me put together something that the world, appreciated the right time. And so a couple of of things I think that are important there. One of them is, I was a consultant in mergers and acquisitions at some point. And what that requires is within three, four weeks, you need to go deep into a new, industry, become an expert, and be able to tell through data and understand the truth through data. Cause the other side is actually going to lie to you and they're going to try to, misrepresent the story so that you pay more or than you want or the company sold for less than it needs to.

And so, that creates a massive amount of a huge bar on the type of analysis that you need to have. So I think that was a key experience. And another key experiences is I created, around 10 years ago, a few viral apps, viral applications on Facebook that got to 20 million users together. And so I experienced, daily exponential growth and things of growing. Something like that grows like five-10% a day. It is not something that most people have experienced and I was lucky to have these types of experience that me positioned to put together these arguments and putting the data first. Right. And every so every argument that I have has the data has the sources because of my M and A experience. And I was able to have a reasonable modeling or understanding of the situation thanks to my viral experience. So it's just an accident that I was reasonably equipped to put this together and I'm sure I'm definitely not the best one, but it was probably the piece that had the most resonance for also a lot of luck, but also because of the, some of these actors.

Gideon Lichfield: Some people might say, actually one of our previous guests said to understand that virus, you actually understand the Coronavirus and how it spreads, you have to have had experience of exponential growth because most people think in a linear form. And to understand that something that is very, very small now could become huge in a couple of weeks.

Tomas Pueyo: And I want to come back to one of the comments you mentioned before that's crucial. It's on the epidemiologists, there's a lot of them who've got any right, right. And then they might not have had been heard or the voice and, and many of them and all of them are more, better trained than I am. But I think one of the pitfalls that I've seen in some of them is that it is easier to make an analysis in retrospect with all the data. And also an analysis that doesn't have consequences on decision making. I'm working with a PhD in epidemiology right now on looking at the, all the literature on, measures, non-pharmaceutical interventions. We call them, social distancing measures and things like that in the past. And there's a lot of papers. There's not all of them that's, that are actionable as in, Hey, this is the cost and benefit of these measures in these circumstances. Right? And so, if the experience is limited to looking backwards with full information and in a way that doesn't create actionable, insights that is not going to be the right type of experience. Whereas a person that has been in a situation where there's exponential growth and needs to make decisions based on imperfect data, which is my day to day to day job, we'll have a perspective that is more actionable. It might not be as accurate. It definitely is not as accurate. Um, but, but it's definitely more actionable.

Gideon Lichfield: What do you think are the biggest things that have changed in your understanding of this pandemic since you started doing this?

Tomas Pueyo: I think that the, one of the core learnings has been more around human behavior, right? The at the beginning when you start, I started looking at numbers. A lot of it became very, it looked obvious that this was going to be exponential and that it was going to be everywhere. And so when I was talking with people and they were not seeing the same thing, you started making these realization right that, Oh, they're not thinking like you. I think one of the key aspects that is that were are not used to heavy change, heavy daily change, that's not the normal human experience. Most things remain the same. Your communities remain the same, your behavior remains the same, the economy, even if it grows exponentially, over such a long period of time that you don't feel it. And so people are not used to very scary... And working in growth, that is your job. And so, I understand why so many countries were late and some of them are still late on the response cause it's very human to see this data and think linear lines instead of exponentially. I think that's a core one. There's a few, many more that we've learned as we go on the virus itself. Right. But I think that's the entire point of the strategy that I recommend of hammer and the dance where there's so much that we don't know that let's stop everything. Understand this better, prepare better, before we can tackle it.

Gideon Lichfield: Right. So let's talk a little bit about what you are saying in these posts. I mean they're long and we are not going to try to summarize them now, but if I can try to characterize very briefly what they're all doing in the first one you're explaining why we need to do social distancing. You're explaining why the virus can spread so fast. And how it can, how it can quickly become very big. In the second, the hammer and the dance and you explain why, the way to defeat it is to go through very strict measures of social distancing at first to bring the level to flatten the curve, as people say, to bring the level right down and then until we find a vaccine or a treatment, we have to maintain the, as it will maintain the foot on the brake for a while. Keep in what you call the dance between strict measures. And a little bit of release to let to let people out, but you have to kind of maintain that equilibrium for awhile until we actually have a treatment. So totally what you're saying now in the third one that just arrived, what do you think is your main point here?

Tomas Pueyo: The third one is focused on the US and the key aspect here is most countries have had a centralized response, which in times of crisis is critical and the US hasn't, it has left most of the decision making to States and that creates some game theoretical issues. Not only in game theory issues such as, for example, they have they are bidding against each other for the same ventilators and the same masks and things like that. And like you don't want that, that's an obvious thing that you want to coordinate at the state level. There's also the fact that none of them have a CDC and so none of them are experts. Uh, you also don't want 50 States to develop their 50 apps for contact tracing or monitoring. You also don't want 50 States to come up with their own tests.

So there's a lot of things that can done at the state level and that should be done at state level, but many of them should be done at the federal level. So I think that's a core point. And then the, it is very connected to the reason why that hasn't happened, which is: a lot of the arguments, especially of the Republican party has uh, and um, and so the, my suggestion on there is it makes, health sense, economic sense and political sense to actually attack it. A health sense, we all know about this, but economic sense, I look back at the experience that we had in the 20th century on epidemiology, and pandemic treatments and going hard against them was better economically than not. And then politically.

Gideon Lichfield: Is that what you mean is taking really strict action at the beginning to control a pandemic so that you can then get back to normal more quickly after?

Tomas Pueyo: That's right. And so I think it's worth actually talking more about the specifics there, especially in analysis on the 1918 flu pandemic where, it arrived from the East coast and went to the West coast. And so the, the East coast cities were surprised a lot of them by this, but the West coast ones learned from it. And so, there is a, trend of East coast cities, taking longer, to take measures and then having shorter measures and then West coast, West coast cities or middle and middle American Western cities, learning from this and making the measures earlier and for longer period of time. And so what they did is controlling for all these aspects of geographic situation and previous economy and all that stuff, which ones of the cities ended up having a better economy afterwards. And there was a strong correlation between the more measures you took and the earlier you took and the better the economy was after that. And so I think the learning from that is not, as much as this is going to happen now, because we just don't know. There's so many things that are different. But the key insight is the only time when we've, this has happened and we've analyzed it, the result was counterintuitive because you would imagine that not shutting down the economy is the better solution, but in fact it was the opposite. So with that type of data that's something to very serious consider really seriously consider for taking measures right now. So that's on the economic side. And then I can also talk about the political side, but you were about to say something.

Gideon Lichfield: No, no. Go ahead.

Tomas Pueyo: I was waiting for somebody to make that argument, but I didn't see it. So I ended up doing it myself. It's just the point that, well mm, the coronavirus hits older people more and older people vote in the US more for the Republican party. Right? So if you put these two things together, you end up, doing back of the envelope numbers where... voters from 2016 where are 30%, three zero, 30% more likely to die from the coronavirus than democratic voters of Clinton.

Gideon Lichfield: Wait, explain that figure little- you mean that. So that's 30% more likely to die, or 30% more of them are likely?

Tomas Pueyo: They are 30% more likely to die. But I think your point is an interesting one. So what does that mean for politics? And so we looked at the per state impact that could have, it could have had and just that effect without secondary effects, just that effect would have, would have, reduced the gap between Clinton and Trump by up to 30% in some states. I think it's specifically been seen in Pennsylvania, 30% of the gap between Trump and Clinton would have been erased just by the death of voters. And then if you add up the votes of their partners or their family and things like that, then my guess is it would've been much higher.

Gideon Lichfield: Right. So I'll just remind everybody who's on the call and listening on Facebook and YouTube you can throw questions to us in the chat. If you're on here on Zoom or in the comments in Facebook and YouTube, we will see them and I will put them to him. Continuing on what you just said. So what's also interesting to me about your third post is you are going, I think further out on a limb. In other words, you're doing political analysis and economic analysis. Do you worry about going as it were, extending yourself too far? I know you, you're doing the kind of analysis that I think people want and that people are talking about and what I haven't seen other people doing those sorts of projections, for instance, for what it means for the balance of power of voting in the States.

Tomas Pueyo: I think I stretched myself too far four weeks ago when I started, wrote the first one. So it's like, there's no way back from this. But I think your point is valid and it's very, very, it's something that I think a lot about. But in slightly different terms. My role is not one of telling people what to do. My role is one of looking at the data and proposing the data to decision makers for them to actually make the decisions. And, and I was very wary of getting into politics, especially in the US because of the provide polarization and the fact that I think it's crucial that, my message is, remains neutral, because otherwise you alienate half of the population and hence half of the, half of the decision making. So, um, I asked for feedback from both sides, a lot of it's to make sure that it was helpful. And I took risks here even more than in my previous post. But I think it's important because, right now official numbers are over 200,000 cases. It's probably the true number of cases, probably above 1 million, probably several million. And so when you have that growth exponentially across all states, across most states, and the measures are not being taken, you are talking about hundreds of thousands if not millions of deaths. And so, we're talking before about a change in the mindset of change, and changing based on new data. In a normal setting, I would never put something out like this because you have time to make the proper arguments for people to debate this, but we don't have that time today. Every day that counts, and so you are forced. It is an imperative for you to go out with what you know, if you think it can help people make decisions.

Gideon Lichfield: Right. I think that is one of the things actually, that people still do not turn, like you say, every day counts because when an epidemic is growing like this, the logic of exponential growth is that delaying something by two days to make the problem considerably harder than it was even in just 48 hours. Um, it coming back, since you're, you know, you're in some ways the ultimate European, I mean I think French, Spanish, right? So when you see the way that Europe responded with every country taking its own approach, closing their borders. I mean EU kind of nowhere to be seen almost and this in this whole process and now you're looking at what the US is doing and seeing this, again, very fragmented decision-making among the different states. Do you feel like Europe should have been more unified in its approach or not?

Tomas Pueyo: So it's very, it's a fascinating comparison. I love it. And in fact, I think it highlights the downside of the US approach. So in Europe, the authorities clear the sovereignty is clear. It is at the national level. And then you have the EU supernational structure, but the authority is at the national level. And so that makes it very clear who needs to make the decisions. Each country has their own CDC and each country then is in a position to centralize all the decision making and very, very crucially close borders, which many European countries have done. That that is not the situation in the US where unfortunately the level of sovereignty wasn't very clear and the federal level has actually, given more sovereignty to the states to handle. And the best example is the one I mentioned before, if every state is buying their own ventilators and masks, then every state is bidding against each other for them.

And all of them are going to be more. And the ones who are going to get them are the rich States. The poor States are the ones who are going to suffer from this. And so this is an obvious area where the government at the federal level has to step in but hasn't. And the other example is the borders, right? So far, I've looked at this and I haven't seen country wide border sealing between States. But it is what's happening in the European union and it is important to limit between different States. And you can imagine that right now, Alaska is doing an amazing job and Texas is not taking as many measures. So what happens if Alaska controls these at high economic cost, but then Texans are flying to Alaska and seeding the epidemic again.

So Alaska will not want to have this and they are going to close borders with Texas, then the Texans might go to Nevada and then go from there or to wherever. And then from there fly to Alaska. So then Alaska is forced to close to seal those borders? Right? And so that is an example of you want that mandated at the federal level because otherwise you're going to start having these kinds of, travel bands and border sealing happening, which is not something you want to do at the, at the state level.

Gideon Lichfield: Right. Well, this is one of the, one of the things that I think also really sticks out in the latest post you write, which is you look at, you look at the how things are playing out state by state. And you point out that I think 33 of the 50 States in the US now are already past the level of deaths that Hubei province was when it shut down. And nowhere in the United States is imposing the kind of travel ban the travel measures, the travel restrictions that China did. And the projections that the white house put out a couple of days ago where they said a hundred between a hundred thousand and 240,000 people could die. Those projections, which I think were coming from the university of Washington, we're also assuming that every state would impose very strict social distancing measures including travel restrictions in the coming week. So, I don't know even what the question is here. I suppose the question is why is we've seen this now play out everywhere and we already know how the dominoes fall. Why are we still so slow to take action here. Well, or put it another way, do you think that the US is, could do that kind of thing as long as they did in China? Now, I'm really asking you to speculate. Sorry. Oh, you're on mute. You're on mute. Okay. Hold on. We have accidentally muted Tomas.

Tomas Pueyo: Yea, I don't know what happened. When I looked at the back of the envelope numbers of mitigation strategy, meaning of basically letting these run, the range goes from 500,000 people to over 10 million. And the range is very big because we don't know so many things about the virus. Right? But, just to put it in context 500,000 people is around the order of how many people who die of cancer every year in the United States. So it is even in the lower range, if you let these run, the impact is dramatic. And you're, the hope by getting these numbers is that we would take measures and then the death rate would be a much lower. Unfortunately, we haven't. And I think the Washington, I've heard about these projections. I haven't looked at the detail. They sound reasonable to me.

I think when I did back of the envelope numbers of two days ago, it sounded like 50,000 deaths was at the very least, it would be in the United States this year if everything went perfectly. So a hundred, 200,000 is reasonable. And to your point, yes. Again, that's if you take measures, but there's a pattern of not taking the measures. And it is true for States on both sides, both Democratic and Republican. But, it's more true for usually rural, States, which makes sense because they're impacted later, but they will be impacted the same and there's also a bit more, this has also been more true for Republican versus Democratic States. And so I think to your point, yes, the number of deaths is going to be staggering much more than we need it to, but still, there's so much more that can happen with every day that we don't take the measures that we need to take.

Gideon Lichfield: Right. I have another question from one of the members of the audience. Have you thought, they ask, have you thought about doing an analysis of the impact of covid on different socioeconomic groups? What do you think the benefit of that information could be?

Tomas Pueyo: I'm guessing what the question is trying to achieve and I guess is the social distancing mandates hurt some groups more than others. Right. And so there's a couple of things I think that are worth noting here. First, when we as a society are taking these measures, in effect, what we're doing is a transfer of wealth from the entire society to older people. Because what we're doing is basically saving old people's lives, which have a certain, amount of money is paying for that. And so I think that that is a transfer that's, is happening and that we, that's what I'm proposing to do, but we need to be conscious that that's what in effect we're doing as a society. Then, that's one aspect. The obvious aspect that most people are going to, are thinking about is the fact that, a lot of the people who have jobs that allow them to work from home, are usually white collar workers that have a higher income than the people who are suffering from these the most, which are going to be blue collar service workers, who can't do their job.

I think for me the same way I was as I was saying that, what we're doing is a transfer from all of society to older people. The same should be happening for these people who are more impacted. If you have a $2,000 income for example, that you give to people, let's say, let's imagine, we do a shutdown for the company economy for three months, $2,000, that's $6,000 in total. Imagine that we give that to 20 million people, which is a fair amount. It's probably, but let's assume just 20 million, 6,000, that is what a hundred and $120 billion, right? So that's around 5% of the, of the 2 trillion that were, have been approved. And so the point I'm trying to make here is we're in a crisis where what's right for everybody is what we need to do, but then some people are going to be more hurt than others and we need to is to mutualize that risk. The groups that are suffering that risk, we need to have a transfer, of that wealth from the groups who are not to them. And that's exactly what, these 2 trillion are for. And that's exactly how we should be thinking about national debt.

Gideon Lichfield: Right. We have only a couple of minutes left, so I'm gonna wrap up by asking you, what are you doing next?

Tomas Pueyo: The number one question I'm getting is now what? And this is true, not for the US only, but rather actually more for the rest of the countries. Many of them have applied the hammer and some of them have officially just use those terms of the hammer and dance. And the question is how do we get out of the dance in how long? And the very quick answer is, the hammer had these two functions. One of them was, eliminating stopping the virus and the other one to learn. And so what you get for free is slowing the epidemic or the outbreak. But what you still need to do is the preparation, is the learning. And so some of the things that I'm not seeing most of these countries is, are you testing properly? Are you setting up contact tracing operations? Do you have quarantine laws? Are you in a position that you can enforce quarantine laws? And so there's all these things, these details of what do you do in a hammer period and how do you promote yourself to dance period. And what is that experience going to be like? And then how does that apply to my specific country? That is the number one question I'm getting, and so I want to focus on that.

Gideon Lichfield: Right. And has this experience made you think, well, maybe you don't want to talk about this when your boss is watching, but has this experience made you think about the rest of your career?

Tomas Pueyo: Yeah, I very purposefully, extract myself from all these thoughts because there is a very pressing problem right now that needs to be, that I need to focus on because I think there's some help I can provide. And everything else is a distraction. Not only is it a distraction, but also if I started thinking about this and there was, it would be a misalignment of incentives where what is right for the message might not be what's right for me and I don't want that. And so I think that's the high level view. I also believe that this is the typical pop and drop thing where everybody's crazy about this for a few months and then he's going to die and they'll go back to normal. Maybe my life changes afterwards. I think all of our lives will change regardless, but it maybe does and I'll just leave that experience and see what happens.

Gideon Lichfield: All right, well thank you for coming to speak to us. Thank you for the work you've been doing. I do honestly think even in this era of misinformation and of so many competing views, I think you have contributed to some clarity for some people.

Tomas Pueyo: That's an entire goal. That's the only ambition.

Gideon Lichfield: Right. Thank you Tomas Pueyo and thanks to everybody for listening and join us on the next episode of right radio Corona. Thank you.

Tomas Pueyo: Thank you.